Search Results

Excelsius3D™ is a mobile X-ray system designed for 2D fluoroscopy, 2D digital radiography, and 3D imaging of adult and pediatric patients. The system is indicated for use where a physician benefits from 2D and 3D information on anatomic structures and high contrast objects with high x-ray attenuation such as bony anatomy and metallic objects. Excelsius3D™ images are compatible with image guided systems such as ExcelsiusGPS®.

Excelsius3D™ is a mobile X-ray imaging system that combines 2D fluoroscopy, 2D digital radiography, and 3D imaging into one unit for acquisition and visualization of 2D or 3D images of patient anatomy. Excelsius3D™ provides real-time image capture, post-capture processing and visualization, and DICOM storage of images.

The provided text describes the Excelsius3D™ device and mentions various performance tests conducted. However, it does not detail specific acceptance criteria, reported device performance metrics against those criteria, or the methodology of a study proving the device meets acceptance criteria concerning AI assistance or standalone algorithm performance.

The document focuses on regulatory clearance based on substantial equivalence to predicate devices, referencing general performance testing categories like "Image quality assessment comparison and paired image analysis to predicates" and "Human cadaveric qualitative validation under clinically relevant scenarios." It also lists compliance with various IEC standards.

Therefore, based only on the provided text, I cannot complete the requested tables and descriptions related to specific acceptance criteria, device performance metrics, study details (sample size, data provenance, expert ground truth, adjudication, MRMC, standalone performance, training set details, or ground truth establishment for AI/algorithm performance).

The request seems to be looking for details typically found in a clinical study report or a more detailed performance evaluation section of a 510(k) submission, which are not present in this summary letter.

To answer your request, if this were a hypothetical scenario where the document did contain such information, the structure would be as follows:

Assuming hypothetical data that would be present in a more detailed submission for this kind of device:

1. Table of Acceptance Criteria and Reported Device Performance

(Note: These are examples of common imaging system performance metrics. The actual metrics would be specific to Excelsius3D's intended use and technological features.)

2. Sample Size and Data Provenance

• Test Set Sample Size: [Number of images or cases, e.g., "100 unique phantom scans and 20 cadaveric scans"]
• Data Provenance: [e.g., "Prospective phantom studies and retrospective human cadaveric data acquired at a US-based university medical center."]

3. Number of Experts and Qualifications for Ground Truth

• Number of Experts: [e.g., "3 independent radiologists and 2 orthopedic surgeons."]
• Qualifications of Experts: [e.g., "Each radiologist had >10 years of experience in diagnostic imaging, with sub-specialty training in musculoskeletal imaging. Each orthopedic surgeon had >15 years of experience in spinal surgery, routinely using intraoperative imaging."]

4. Adjudication Method for Test Set

• Adjudication Method: [e.g., "Majority vote (2+1) among the three radiologists for image quality assessment. Any discrepancy was resolved by consensus discussion with all three radiologists present. For cadaveric studies, anatomical ground truth was established by direct measurement during dissection or by a consulting anatomist."]

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

• Was an MRMC study done? [Based on the provided text, a formal MRMC comparative effectiveness study demonstrating human reader improvement with AI assistance is not indicated. The document focuses on the device's technical performance and substantial equivalence to predicates.]
• If yes, effect size of human reader improvement: [If such a study were done, it would quantify the improvement in diagnostic accuracy, confidence, or efficiency when readers use the AI-assisted system versus conventional methods. e.g., "An MRMC study showed a 15% increase in sensitivity for detecting subtle fractures when radiologists used the AI system, while maintaining specificity."]

6. Standalone (Algorithm Only) Performance

• Was standalone performance done? [The provided text does not describe an AI/algorithm component requiring standalone performance evaluation. The device is an image acquisition system. If it had integrated AI for automated detection or measurement, then yes, standalone performance would be applicable.]
• If yes, details: [e.g., "Standalone performance of the automated fracture detection algorithm achieved an F1-score of 0.92 on the test set."]

7. Type of Ground Truth Used

• Type of Ground Truth: [e.g., "For phantom studies, the ground truth was known physical properties and configurations of the phantom. For cadaveric studies, ground truth for anatomical structures and implanted devices was established through direct observation during dissection, and confirmed by an expert anatomist's review of pre- and post-dissection imaging (e.g., high-resolution CT/MRI) and photographic documentation."]

8. Training Set Sample Size

• Sample Size for Training Set: [Not applicable for a purely physical imaging system without a machine learning component requiring a training set. If there were ML components, then "Not explicitly stated in the provided document, but typically thousands to tens of thousands of images would be used for training deep learning models."]

9. How Ground Truth for Training Set Was Established

• How Ground Truth Was Established (Training Set): [Not applicable for a purely physical imaging system. If there were ML components, then "Not explicitly stated in the provided document, but typically through expert annotation (e.g., radiologists, surgeons) following a pre-defined labeling protocol."]

Ask a specific question about this device